Project Summary (Abstract) Glioblastoma (GBM) is the most common and aggressive primary brain tumor. The infiltrative nature of tumor cells makes surgical resection incomplete. Furthermore, recurrence is inevitable despite radiation and temozolomide treatment and patients die within 15 months following diagnosis. GBM are divided into several molecular subtypes based on distinct gene expression profiles, including proneural (PN), mesenchymal (MES), classical (CL). One important reason that current anti-neoplastic therapies fail to provide a durable response in GBM is the adaptive nature of the tumor microenvironment. The most abundant non-neoplastic cell population in the GBM microenvironment is tumor-associated macrophages (TAMs). Of the GBM subtypes, MES expresses the highest levels of TAM-associated genes. Our analysis of TAM numbers revealed that MES GBM also has the highest number of TAMs compared to the other subtypes. When we correlated the high and low expression levels of TAM-associated genes with patient survival in a subtype-specific manner, only PN GBM patients showed a correlation of high expression: short survival and low expression: long survival. PN GBM is also known to be the most resistant to anti-neoplastic cell-specific targeted therapies. To examine TAM-GBM cell interactions, we used genetically engineered immunocompetent mouse models of PDGFB- and NF-1 loss- driven GBM and showed that tumor cells induce TAMs to produce the key pro-inflammatory cytokine IL-1?. TAMs release IL-1?, which binds to the receptor IL-1R1 on tumor cells and leads to activation of IL-1? signaling in PDGFB-driven GBM cells, which leads to i) increased stemness and growth, and ii) increased expression of the monocyte chemoattractant protein (MCP) network (CCL2, CCL7, CCL8, CCL12) in PDGFB- driven GBM cells. Our data showed that loss of IL-1? from the microenvironment resulted in a significant decrease in PDGFB-driven GBM formation and growth compared to wild-type mice in vivo. Based on our data, we hypothesize that TAM interaction with GBM cells is cell type- and subtype-specific and that they have different functions in PN and MES GBM subtypes. In this application, we will determine the detailed mechanism by which PN and MES GBM cells recruit and alter the function of macrophages to create specialized TAMs (Aim 1). To determine the mechanism of IL-1? signaling, the MCP network, and their downstream targets in PDGFB- and NF1 loss-driven GBM using genetically engineered mouse models and human GSC-derived tumors in vivo. Determine the mechanism by which TAMs and IL-1? support the creation and maintenance of the perivascular niche, which provides the proper microenvironment for glioma stem cells ? the treatment-resistant population of glioma (Aim 3). The proposed studies will provide new mechanistic insights into fundamental cellular and molecular biological processes related to TAM? tumor cell interaction in vivo, and will allow for identification of novel potential therapeutic targets to glioblastoma.